Latitudinal variation in tritrophic interactions associated with native and exotic genotypes of Phragmites australis

Despite theoretical expectations, studies of latitudinal gradients in plant-herbivore interactions have produced mixed results. Contradictory findings may be due to methodological flaws such as ignoring the influence of other trophic levels (e.g. predators, parasitoids, and inquilines), sampling few locations or a narrow latitudinal range, and not accounting for differences between plant species or genotypes. We investigated tritrophic interactions between native and invasive genotypes of common reed (Phragmites australis), gall-inducing Lipara spp. (Diptera: Chloropidae – L. pullitarsis, L. rufitarsis and L. similis), and their natural enemies and inquilines at 36 sites along the North American east coast ranging from South Florida (26.6°) to New Brunswick, Canada (48.4°). Contrary to predictions from theory, we found that the proportion of stems infested with Lipara spp. galls increased significantly with latitude and was also significantly higher in the native than exotic genotype. The frequency of inquiline fly co-occurrence in galls was also significantly higher in more northern sites but did not differ between genotypes. Parasitism rate of galls was virtually zero, regardless of Lipara species, site latitude, or P. australis genotype. These results indicate that enemy-release may play a part in the successful invasion of the exotic P. australis genotype and its associated species, and that invasion success may vary over a broad spatial scale. Future study directions include a controlled common garden experiment to examine if the gradient has a genetic basis.